DESCRIPTION: Microsatellites are short, simple sequence repeats that are present in mammals in thousands of copies dispersed throughout the eucaryotic genome. Length variation at microsatellite loci is common, and it has recently been demonstrated that microsatellite length variation is a marker for genetic instability in some cancers. The basis for this length variation is not known, but the two major models include slipped-strand mispairing and unequal recombination. The most common microsatellite, CA/GT, is found in tracts ranging from 6 to 30 repeats. Length variation is uncommon at loci with tracts of fewer than 10 repeats, and appears maximal at loci with tracts of 11 to 17 repeats. Perfect repeat tracts are more variable in length than imperfect repeat tracts. Dr. Farber has developed a system which allows her to screen for changes in lengths of microsatellites in cultured mammalian cells. In her system, microsatellite repeats which are not multiples of three bps are fused to a selectable marker (neo) so that the gene is out of frame; the correct reading frame can be restored by changes in the number of repeats. Restoration of reading frame results in resistance to G418 (a neomycin analogue). Ability to select these mutations provide a sensitive method for detection of these changes. Successful experiments carried out using a poly(CA) repeat in CAK mouse cell lines have been published. In these experiments, reversion rates were 100 times higher in cells carrying the repeat than in control cells. This system can be used to address a number of questions about mutations at microsatellite loci. This vector can theoretically be used in any cultured cell line, and microsatellites of different lengths and compositions can be tested. The specific aims of the proposed research are: to test the hypothesis that the extent of polymorphism for a given microsatellite is directly related to the mutation rate for a sequence of that length; to test hypotheses regarding the relationship between sequence composition and mutation rate; to compare microsatellite mutation rates of normal and cancer cell lines; to determine the stage in development of a malignancy that microsatellite instability is first observed; to study the effects of genes which have been implicated in genomic instability (e.g. p53 and Werner syndrome); and to develop a test for frameshift due of mutagenic compounds.